John S. Rowan

Equifinality and uncertainty in physically based soil erosion models : application of the GLUE methodology to WEPP-the water erosion prediction project-for sites in the UK and USA.

Despite the wealth of soil erosion models available for the prediction of both runoff and soil loss at a variety of scales, little quantification is made of uncertainty and error associated with model output. This in part reflects the need to produce unequivocal or optimal results for the end user, which will often be an unrealistic goal. This paper presents a conceptually simple methodology, Generalized Likelihood Uncertainty Estimation (GLUE), for assessing the degree of uncertainty surrounding output from a physically based soil erosion model, the Water Erosion Prediction Project (WEPP). The ability not only to be explicit about model error but also to evaluate future improvements in parameter estimation, observed data or scientific understanding is demonstrated. This approach is applied to two sets of soil loss/runoff plot replicates, one in the UK and one in the USA. Although it is demonstrated that observations can be largely captured within uncertainty bounds, results indicate that these uncertainty bounds are often wide, reflecting the need to qualify results that derive from optimum parameter sets, and to accept the concept of equifinality within soil erosion models. Attention is brought to the problem of under-prediction of large events/over-prediction of small events, as an area where model improvements could be made, specifically in the case of relatively dry years. Finally it is proposed that such a technique of model evaluation be employed more widely within the discipline so as to aid the interpretation and understanding of complex model output.

Multi-parameter fingerprinting of sediment deposition in a small gullied catchment in SE Australia

The determination of relative contributions of potential sediment sources is an important step in the development of management strategies to combat soil erosion. In a 1.2 km2 gullied catchment in southeastern New South Wales, multi-parameter fingerprinting of sediment deposited in successive downstream pools has identified gully walls as the dominant sediment source when the grazed pasture surface was the only other potential source. The median fractional contributions remained relatively steady in the successive downstream pools, with the gully walls responsible for between 90% and 98% of the pool sediment. This result was achieved despite the ratio of the source areas varying considerably between successive nested subareas. Reliability bounds on the predictions, accounting for limited sampling of sources, were well constrained and varied between 5.4% and 13.8%. Downstream of an unsealed road crossing, sediment from the road source dominated the pool sediments such that contributions from the pasture surface and gully sources could not be determined.

Approaching the physical-biological interface in rivers: a review of methods for ecological evaluation of flow regimes

New European legislation known as the Water Framework Directive (WFD) challenges catchment hydrologists and freshwater biologists to quantify the risk of damage to the organic communities of rivers that arises from anthropogenic distortion of the natural flow regime. Here, we take the first step towards this goal by collecting together relevant information from the two disciplines. An extensive biological literature is examined for insights into the ways in which the species and communities associated with rivers might change when the flow regime is altered. From the hydrological literature, the indicators of flow regime and flow regime change that are pertinent to ecology are described, and consideration is given to means of deriving flow regime data for ungauged river reaches. Attempts to combine hydrology and ecology in classifying rivers and in setting flow objectives to favour biota are then reviewed, together with integrated approaches to river management that aim to promote ecological quality. A significant scale disparity is noted between the disciplines, hydrology being studied at catchment, subcatchment and reach scales, and biology generally at local level. Nonetheless, both yield methods with potential applications in aspects of WFD implementation. The approach with most appeal for general risk assessment is based on the concept of hydrological alteration. This technique employs flow regime variables selected for their importance to aquatic and riparian ecology, and quantifies deviations from the natural values of these variables at reach scale. For WFD purposes, calibration of the scale of hydrological alteration in terms of risk to ecological status is desirable. In this, priority should be given to identification of the level of hydrological alteration that corresponds to the division between good and moderate ecological status.

MIRSED towards an MIR approach to modelling hillslope soil erosion at the national scale.

This paper reports a new methodology for assessing regional and national patterns of hillslope scale soil erosion rates in the UK using a MIR (minimum information requirement) version of WEPP (Water Erosion Prediction Project) known as MIRSED. WEPP is parameterised using a national coverage, environmental database containing topographic, soil, land management and climate variables for all hillslopes within each grid cell to be modelled. The MIRSED matrix summarises the behaviour of WEPP in a multi-dimensional parameter space, allowing results to be queried using a subset of key, spatially variable parameters to produce an averaged hillslope soil erosion response from each 1 km2 grid cell. The approach is demonstrated for the Great Ouse catchment, Cambridgeshire, UK and highlights highest hillslope erosion rates of 2.2 t ha−1 year−1 associated with steepest slopes, erodible soils and management practices that leave the soil exposed for critical times of the year. A mean soil erosion rate of 0.4 t ha−1 year−1 is predicted from hillslopes across the catchment which compares well with observed data collated at different scales, using contrasting measurement techniques.

Storm Event Suspended Sediment-Discharge Hysteresis and Controls in Agricultural Watersheds: Implications for Watershed Scale Sediment Management

Within agricultural watersheds suspended sediment-discharge hysteresis during storm events is commonly used to indicate dominant sediment sources and pathways. However, availability of high-resolution data, qualitative metrics, longevity of records, and simultaneous multiwatershed analyses has limited the efficacy of hysteresis as a sediment management tool. This two year study utilizes a quantitative hysteresis index from high-resolution suspended sediment and discharge data to assess fluctuations in sediment source location, delivery mechanisms and export efficiency in three intensively farmed watersheds during events over time. Flow-weighted event sediment export was further considered using multivariate techniques to delineate rainfall, stream hydrology, and antecedent moisture controls on sediment origins. Watersheds with low permeability (moderately- or poorly drained soils) with good surface hydrological connectivity, therefore, had contrasting hysteresis due to source location (hillslope versus channel bank). The well-drained watershed with reduced connectivity exported less sediment but, when watershed connectivity was established, the largest event sediment load of all watersheds occurred. Event sediment export was elevated in arable watersheds when low groundcover was coupled with high connectivity, whereas in the grassland watershed, export was attributed to wetter weather only. Hysteresis analysis successfully indicated contrasting seasonality, connectivity and source availability and is a useful tool to identify watershed specific sediment management practices.

A multi-method study of bedload provenance and transport pathways in an estuarine channel

Abstract Bedload transport pathways and provenance in the upper reaches of the Tay Estuary, Scotland have been investigated using a combination of three methods, analysis of bedform geometry and asymmetry using echo-sounding and side-scan sonar, the ‘McLaren Model’ of deducing sediment transport from grain size distributions and magnetic susceptibility measurements. The 13.5 km long study reach has been shown to be a complex mixing zone in which bedform asymmetry varies both temporally and spatially. Both grain size transport trends and magnetic susceptibility data suggest that the area is dominated by marine derived bedload but, to the south of the main channel axis, the contribution from fluvial input increases. A simple mixing model based on optimised linear programming has shown the relative contribution made to estuary bed sediments from marine and fluvial sources. With increased distance upstream the role of fluvial sources becomes more pronounced peaking at 24%. Together the results clearly illustrate the dominance of marine-derived bedload in the study area.

Long-term sediment yield in Crombie Reservoir catchment, Angus; and its regional significance within the Midland Valley of Scotland

Abstract The drawdown of Crombie Reservoir in November 2001 afforded the opportunity to examine the exposed sediments trapped since impoundment in 1868. Direct measurements of infill depth enabled an isopachyte map to be produced. Gravimetric conversion using measured bulk densities and a trap efficiency term indicated a long-term catchment sediment yield of 59.1 t km−2 year−1. Core stratigraphy analysis indicated that sediments were dark brown/black cohesive silty-muds with multiple sandy sub-units, representing a combination of discrete flood events and previous drawdown surfaces. Dating, constrained by mineral magnetic and 137Cs analysis, indicated that sedimentation rates have varied from 0.2 to 0.8 g cm−2 year−1, corresponding to a four-fold variation in catchment sediment yield (approximately 20–93 t km−2 year−1), most likely controlled by extensive conversion of moorland to woodland, and post-World War II agricultural expansion. The Crombie investigation is combined with other reservoir sedimentation surveys within the Midland Valley of Scotland. Area-specific sediment yields (t km−2 year−1) evidence a weak, though statistically significant (p > 0.05), positive correlation with catchment area (km2). The increase in area-specific yield with catchment area contradicts the decline, which is generally expected, and is taken to reflect the significance of channel erosion within water supply basins featuring mainly natural and semi-natural vegetation cover and low-intensity land management practices. With stable slopes channel erosion dominates and area-specific sediment yield increases downstream due to greater entrainment and transport potential. The high degree of scatter in the Midland Valley database reflects significant variations in the extent of land-use change and the local importance of agricultural improvements and afforestation practices.

Bayesian sediment fingerprinting provides a robust tool for environmental forensic geoscience applications

Abstract Sediment fingerprinting is an approach for the quantitative determination of sediment provenance (both spatial sources and types of sediment supply) over a range of temporal and spatial scales. Though widely adopted, studies often vary in their attention to the underlying assumptions and in their treatment of modelling uncertainty. A Bayesian approach to the multivariate problem of ‘unmixing’ sediment sources is reported, showing the significance of source group variability and source group sampling density to the accuracy of model output. The model produces results as median source group contributory coefficients (and associated 95% quantiles). The model was applied to environmental data obtained from selected soil erosion studies reported within the peer-reviewed literature. Good correspondence (r2=0.89) between reported mean source group contributory coefficients and median values were found when recalculated using the Bayesian analysis. However, confidence levels are highly variable, ranging from 2% to 97%. The robustness of any unmixing solution depends on factors such as the number of samples, the number of source groups and the variance of source group properties. It is concluded that ‘forensic-style’ investigations must recognize these uncertainties and be appropriately resourced to achieve tolerable accuracy and precision. The discussion considers additional confounding factors such as non-conservative tracer behaviour and enrichment/depletion during the sediment delivery process.

Reconstructing historic reservoir sedimentation rates using data-based mechanistic modelling

This paper reports an application of data-based mechanistic (DBM) modelling to the study of reservoir sedimentation. A detailed monitoring programme in the catchment of the Wyresdale Park reservoir was used to calibrate a two component DBM sedimentation model. The first component was a non-linear rainfall to suspended sediment load (SSL, kg s−1) model, the second dealt with sediment routing and the trap efficiency of the reservoir. Daily precipitation data for the period 1911–1996 were used to simulate the sedimentation history of the reservoir. The synthetic accretion sequence evidences the effects of climatic forcing and was compared to lake-bed sediment cores independently dated using 137Cs. The synthetic stack showed general agreement with the observed accretion data. Departures in model performance most likely reflect non-stationarity in the system due to local changes in land use and reservoir regulation.

A 2-D Reservoir Routing Model: Sedimentation History of Abbeystead Reservoir, U.K

A physically based two-dimensional model is applied to the case-study site of Abbeystead Reservoir, U.K. The model, developed fordensity currents, solves the Navier-Stokes equation coupled to ageneral sediment transport equation. Water flow and sediment motion are determined by reference to dimensionless Reynolds andRichardson numbers. An additional dimensionless parameter, πa,determines suspended and/or bedload sediment transport rates. Theperformance of the model is considered in relation to data obtained from repeated bathymetric surveys (1876–1991) for Abbeystead Reservoir, which provides an exceptionally detailed long-term record of sedimentation rates and deposition patterns.The suitability of the adopted two-dimensional scheme is relatedto the local morphology of the reservoir and the time-scale of the delta propagation process. First results show good agreementbetween the numerical simulations and the field data. The modelis able to reproduce both the dynamics of delta growth and the quasi-equilibrium delta form reached as the infilling process nears completion.